Contribution of RNA and nonhistone proteins to the circular dichroism

Effect of cyclic AMP on chromatin-bound protein kinases in WI-38 fibroblasts stimulated to proliferate. Bernd M. Bombik , Renato Baserga. Biochimica e...
0 downloads 0 Views 1MB Size
HJE1.M

A 1 D HLAhC

The Contribution of RNA and Non-Histone Proteins to the Circular Dichroism Spectrum of Chromatin? Rex P. Hjelm, Jr.,i and R u Chih C . Huang*

ABSTRACT: This paper is an investigation of the contribution of low salt extractable R N A and non-histone proteins to the circular dichroism of chromatin. Circular dichroism (CD) of chromatin above 250 nm is due mainly to D N A and is different from that of D N A free in solution. In addition, to a smaller extent, we find that low salt extractable

R N A and/or non-histone protein side chain chromophores contribute significantly to the spectra in this region and account for the major differences observed among the C D spectra of chromatins isolated from the five tissues studied: pig cerebellum, myeloma, calf thymus, chick embryo brain, and chick erythrocytes.

s t u d i e s on isolated chromatin find it to consist of D N A , protein, and small amounts of R N A (Bonner et al., 1968; Stellwagen and Cole, 1969; Hearst and Botchan, 1970; Georgiev, 1969; DeLange and Smith, 1971; Elgin et al., 1971; Huang and Hjelm, 1974). The proteins consist of histones, a discrete group of highly basic, small proteins which are intimately associated with the DNA; and the non-histone proteins, a class of highly heterogeneous proteins which includes enzymes, structural proteins, and genetic regulators. Determination of the interaction and conformation of chromatin components is a problem of great importance. Its resolution is key to an understanding of the basis of chromosome structure and function in higher plants and animals. This and two related papers (Hjelm and Huang, 1974a,b) deal with the circular dichroism (CD) of chromatins isolated from tissues differing in nuclear synthetic activities. Of these, three produce R N A and DNA, calf thymus, 1 1-day-old chick embryo brain, and myeloma K41 (mucine solid tumor); one, pig cerebellum-like other adult nervous tissues (Deluca et al., 1953; Bendick et al., 1953; Leblond and Walker, 1956; Koenig, 1958-produces only R N A ; and one, chicken erythrocyte, is completely repressed in nuclear synthetic activity (Williams, 1971; Neelin et al., 1964). In addition, the in situ nucleochromatin of avian n u cleated erythrocytes is completely condensed (Vidali et al., 1973). unlike the nucleochromatins of the other tissues studied. These chromatins also differ in the relative amounts of the various chromatin components; specifically non-histone proteins and R N A . In our initial observations on some of these chromatins we noted differences among their C D in the positive ellipticity region above 260 nm (Hjelm and Huang. 1972). Similar variations have recently been reported by Lin et al. (1974) in chromatins isolated from normal diploid and SV40 transformed human fibroblasts. These differences are

either due to variations in the secondary structure of chromatin DNA, or to the trivial contribution of R N A (as suggested by Simpson and Sober, 1970) and/or protein sidechain chromophores. Resolution of this question has considerable intrinsic value and is important to our studies on the roles of histone and non-histone proteins in determining the secondary structure of D N A in chromatin (Hjelm and Huang, 1974a). In addition, a study of the possible contribution of other chromophores besides D N A to this region may give information on the structure of the R N A and the cystine and aromatic-containing proteins i n chromatin. In this paper we report that the R K A and/or protein side-chain chromophores contribute significantly to the spectrum above 260 nm of chromatins; this contribution accounts for the major differences observed in the C D of chromatins in this region of the spectrum.

7 From the Department of Biolog). The Johns Hopkins Universit), Baltimore, Maryland 21218. Received June 7 , 1974. This work was supported in part by Grants CA 13953 and IROl HD 08185 from the National Institutes of Health. R.P.H. was a Predoctoral Fellow of the hational Science Foundation (1968-1972) and a Trainee of the National Institutes of Health (1972-1973). $ Present address: Biophysics Laboratories. Department of Physic>, Portsmouth Polytechnic, Gun House, Hampshire Terrace. Portsmouth POI 20G, England.

1682

BIOCHEMISTRY, VOL

14,

10.8

. 1975

Experimental Section

Materials Calf thymus (Henry W . Stapf, Inc.) and pig cerebellum (Esskay Quality Meat Company) are obtained immediately after slaughter and placed on ice. Blood from 6-month-old hens is collected from a neck wound and placed in a 10% (w/v) solution of trisodium citrate (100 mi of solution/ 1000 ml of blood; Murray et al., 1968). Myeloma tumors are cultured in vivo in Balb/c mice, and are provided by the courtesy of Mrs. Susan Ayang and Dr. Janet Stavnezer. Chick brains are obtained from 1 I-day-old embryos. All materials are used immediately after obtention, with the exception of calf thymus which is often stored at -20' until use.

Met hods Isolation of, Nuclei. Pig cerebellum and chick embryo nuclei are isolated by the method of Shaw and Huang (1 970). Nuclei from chicken erythrocytes are prepared according to the procedure outlined by Murray et al. (1968). Nuclei from myeloma are prepared by the method of E. Stavnezer (personal communication). According to this procedure, 4-5 g of minced tissue is made into a 10% homogenate (w/v) in 0.32 M sucrose, 5 m M MgC12, 10 m M Tris (pH 7.9), and 0.5% Triton X-100 (homogenization medium) by ten passes in a Teflon-glass homogenizer. The homogenate is filtered through eight layers of cheesecloth,

VARIATION

OF CHROMATIN CIRCULAR DICHROISM ABOVE 250 N M

then centrifuged a t 3500 rpm for 15 min in a Sorvall SS-34 rotor. The pellet is resuspended into fresh homogenization medium without Triton X-100 (1 ml/g of starting material). To this mixture is added three volumes of 2.3 M sucrose, 5 m M MgC12, and 10 m M Tris (pH 7.9). The suspension is centrifuged at 16000 rpm for 45 min in an SS-34 rotor. The pellet contains the purified nuclei. Preparation of Chromatin. Chromatin is isolated by the method of Huang and Huang (1969). Except for calf thymus, where finely chopped whole tissue is used, the starting material is isolated nuclei (Shaw and Huang, 1970). Chromatin isolation and all subsequent procedures are done a t 4O, unless otherwise noted. Histone degradation can occur during the isolation of calf thymus chromatin, but can be prevented by the addition of sodium bisulfite (0.05 M , p H 8.0) to the salineEDTA (Bartley and Chalkley, 1970). N o histone degradation is observed in the other chromatins, and the standard saline-EDTA solution is used in these cases. LOW Salt Extraction of Chromatin. Chromatin is extracted in solutions of saline-EDTA or of 5 m M phosphate and NaCl ranging in ionic strength from 0.2 to 0.35. To unsheared chromatin (1 5-20 OD260/ml) ice-cold 0.2M sodium phosphate ( p H 6.8) is slowly added with constant stirring, until a concentration of 5 m M is reached. Cold 5 M NaCl is then added in the same manner until the chromatin suspension is a t the desired ionic strength. During this procedure the chromatin is kept cold in an ice bucket. The precipitated chromatin is stirred for a few minutes, then pelleted a t 5000 rpm (SS-34 rotor) for 10 min. The pellets are made into nucleoprotein gels by swelling into solutions of decreasing Tris buffer (pH 8.0) as in the chromatin isolation procedure. The extracted chromatin is termed nucleoprotein low salt. The extract containing R N A and non-histone proteins and some D N A is termed low salt extract. Quantitative Determination of Chromatin Components. Protein determinations are made by the method of Lowry et a]. (1951) and are made directly on the various chromatins and depleted nucleoproteins. Determination of the R N A content of each sample is done by first hydrolyzing the sample with 0.3 M K O H for 18 hr a t 37O, followed by the precipitation of the KOH, unhydrolyzed DNA, and protein with 8% perchloric acid. The amount of soluble ribonucleotides is then determined by the orcinol reaction (Schneider, 1957). Diphenylamines (Burton, 1956) for D N A are made after hydrolysis in hot 5% perchloric acid (70°, 10 min). Isolation of RNA for Electrophoresis and Circular Dichroism. R N A is extracted from the low salt extract with buffer-saturated phenol containing 0.5% sodium dodecyl sulfate (SDS).' The phenol-extracted R N A is dissolved in 0.01 M sodium acetate (pH 4.7) and applied to a small DEAE-cellulose column equilibrated in the same buffer. The column is then washed with 0.3 M NaC1, 10 m M MgC12, and 0.01 M sodium acetate ( p H 4.7). Elution of low molecular weight R N A from the column is effected by 1.O M NaCl. Sixty percent of the OD260 loaded on the column is in this fraction. The R N A is dialyzed against l o v 2 M NaCl for C D and against distilled water for acrylamide gel electrophoresis. Analysis of Chromosomal Proteins. Chromosomal proteins are analyzed by SDS and urea acrylamide gel electrophoresis. Low salt extract samples a r e dialyzed against distilled water, then lyophilized. Total chromosomal proteins Abbreviation used is: SDS,sodium dodecyl sulfate

and proteins from nucleoprotein low salt samples are dissociated from D N A by 7 M urea-2 M NaCl (Shaw and Huang, 1970), followed by centrifugation a t 30,000 rpm for 36 hr a t 4O in a Spinco 30 rotor. The protein-containing supernatant is dialyzed and lyophilized. The dried samples are dissolved in either 4 M urea-0.9 M acetic acid, or 0.1% SDS. Samples are reduced by incubation in 0.1 M 0-mercaptoethanol a t 37' for 1 hr. The samples in urea-acetic acid are applied to 15% polyacrylamide-urea gels (after Panyim and Chalkley, 1969), and electrophoresed a t 100 V for 4 hr. The gels are stained with 0.1% Buffalo Blue-Black (in 7% acetic acid) for 4 hr. Destaining is carried out by diffusion into 7% acetic acid. Samples dissolved in S D S are applied to 8.75% polyacrylamide-SDS gels (after Laemmli, 1970). Electrophoresis through the running gel is done a t 2 mA/gel. Bromophenol Blue is used as a marker. T h e SDS gels are stained and destained according to the protocol outlined by Fairbanks et al. (1971); Coomassie Brilliant Blue is used to stain the proteins in this procedure. Measurement of Circular Dichroism. Chromatins and depleted nucleoproteins are sheared (Waring Blendor, 80 V variac 1 min) then dialyzed against 0.7 m M sodium phosphate (pH 6.8). Samples of low salt extract are measured in the solutions in which they are extracted. Concentrations of chromatin or nucleoproteins are adjusted to between 1.0 and 1.7 O D a t 260 nm. Unless otherwise stated, the C D spectra are measured a t 4O. C D spectra are recorded on a Cary 60 spectropolarimeter fitted with a 6003 C D attachment; IO-mm path-length cells are used. Uv spectra are measured by a Cary 14 spectrometer or a Beckman D U spectrometer. Calculations. C D data are collected a t intervals of 2.5 nm from the chart readout of the Cary 60. The data are reported as molar ellipticity [(deg cm2)/dmol] and are calculated from the data by the equation: [e] = (M/IOcl)B, where [e] is the molar ellipticity, M is the average molecular weight of a nucleotide or amino acid residue, c is the concentration of DNA, R N A , or protein in g/ml, 1 is the path length of the cell in centimeters, and 0 is the measured ellipticity of the sample. M is 309 for DNA, 327 for R N A , and 1 15 for protein. All data are presented plus and minus standard deviation. Reduction of the C D data is carried out by Basic and Fortran I1 programs run on a Hewlett-Packard 2100 computer. Results and Discussion

Circular Dichroism above 250 nm of Chromatins. Consistent with the observations of Shih and Fasman (1970) the C D spectra above 250 nm of the various chromatins are different from that of D N A (Figure 1). It is also apparent that the C D of the various chromatins are different in some cases, especially in the region about 273 nm where chromatins isolated from pig cerebellum, chick embryo brain, and myeloma have a peak (Figure 1) of about 6000 (deg cm2)/ dmol of D N A (Table I); chicken erythrocyte and calf thymus chromatin demonstrate only a shoulder in this region (Figure l ) , 4000-5000 (deg cm2)/dmol of D N A in amplitude (Table I). The spectral differences between the two sets of chromatins are better shown by the ratio of ellipticities a t 273 and 283 nm (Table I). Pig cerebellum, chick embryo brain, and myeloma chromatins have 273/283 ratio (about 1.16) which is significantly different from the ratio (approximately 1.O) for chicken erythrocyte and calf thymus chromatins. BIOCHEMISTRY, VOL.

14,

NO.

8, 1975

1683

H J E L M A h D HUANG

Table 11: Protein to DNA Mass Ratios of Chromatins and Chromatins Dissociated by Solutions of Low Ionic Strength. Source

Wavelength ( n m ) F I G U R E 1: Circular dichroic spectra of various chromatins. (-)

Chick embryo brain; (- - .) pig cerebellum and myeloma; (- -) chicken erythrocyte; -) calf thymus; (- - -) calf thymus DNA. The spectra are averages of several determinations as indicated in Table 11, but retain the characteristics of the individual spectra. Molar ellipticities are expressed as (deg cm2)/dmol of DNA.

.

(a

Table I : Ellipticities at 273 nm and 283 n m of Chromatins and Nucleoproteins and Their Spectral Ratios. Chroma tin Source

273nm

Chick embryo brain

6220

283nm

620 5100 i 480 1.19 t 0.09b (18 F Pig cerebellum 5800 i 470 4670 k 480 1.16 i 0.08b (17) Myeloma 5 7 5 0 i 630 4 6 2 0 2 660 1.15 i: 0.03b (9) Calf thymus 4000 i 480 4650 i 610 0.93 + 0.08b (7) Chicken 4 7 9 0 t 290 5 3 1 0 i 570 erythrocyte 0.86 I 0.08b (6) k

NP Low Salt0 273nm 4400

Chick embryo brain Protein/DNA Fraction of protein0 No. of determinations Pig cerebellum Pro tein/DNA Fraction of protein No. of determinations Myeloma Protein/DNA Fraction of protein No. of determinations Calf thymus ProteinjDN.4 Fraction of protein N o . of determinations Chick erythrocyte Pro tein/DNA Fraction of protein No. of determinations

Chromatin

NP Low Salt

3.3 i 0.3 1 .o (17)

2 . 1 t 0.3 0.66 f 0.07 (6)

3.1 1.0 (17)

_+

0.3

2.6 i 0.2 0.86 k 0.04 (12)

2.3 i 0.4 1.0 (9)

1.9 0.79 (1)

1.9 i 0.3 1 .o (12)

1.9 0.86 (1)

2.2 1.0 (7)

i

0.1

0The fraction of protein remaining on each sample relative to that of the original chromatin sample.

Table 111: RNA to DNA Mass Ratio of Salt-ExtractedChromatin$.a

283nm

560 4550 I490 0.96 i 0.05b (18) 4740 i 600 4270 5 480 1.08 i 0.08b (1 7) 3898 4455 0.87b (1) 3850 4275 0.90b (1) i

Source

CHRb

Sample NP LSC

P LSd

Chick embryo brain Pig cerebellum

0.102 i 0.020 (7)e 0.122 i 0.020 (9) 0.120 i 0.040 (3) 0.044 i 0.008 (3) 0.030 ( 1 )

0.064 i 0.016 (10) 0.084 ? 0.034 (13) 0.080 (1)

0.043 i 0.006

Myeloma Calf thymus Chicken erythrocyte

(8)

0.030 (1)

UNucleoprotein low salt (see Methods). bRatio of ellipticities a t 273 and 283 nm. CNumber of determinations given in parentheses.

URNA determined by the orcinol reaction (Schneider, 1957); DNA determined by the diphenylamine procedure (Burton, 1956). bChromatin. CNucleoprotein lo& salt (see Methods). dLow salt CYtract. The ratio given here is the amount of RNA in the extract relative t o the total DNA in the chromatin sample from which the extraction is made. eNumber of determinations given in parentheses.

The difference is significant to greater than 0.99 by Student’s t test. Chromatin, like other macromolecular aggregates, scatters light. This can result in C D artifacts (Gordon, 1972; Dorman and Meastre, 1973; Holzwarth et al., 1974), and could account for the observed C D differences between the chromatins. However, all the chromatins have about the same turbidity; the O D 320/OD260 ratio is about 0.05 in most cases. Turbidity of the samples with large amounts of protein can vary a great deal, depending on handling; yet we find no detectable variation in the C D about 283 nm. The spectra are noisier when the samples are turbid, but it appears that this effect does not account for the differences observed. Composition of the Chromatins Studied. The differences observed in the C D about 273 nm of the various chromatins may reflect a variation in D N A secondary structure with different R N A transcription activity, or they may reflect the presence of protein side-chain or R N A chromophores; thus a chemical analysis of the chromatins is necessary. The amounts of R N A and protein relative to D N A ( g g /

gg) for the various chromatins are given in Tables I1 and 111. The R N A content is greatest in those chromatins isolated from tissues having the highest capacities for R N A synthesis (Table 111); these are also the samples demonstrating the higher ellipticities about 273 nm (Figure 1. Table I ) . A similar correlation appears to be present for protein content, with the exception of myeloma chromatin (Table 11). Studies on the histone content of chromatins from a wide variety of plants and animals reveal that the histone/DNA mass ratio is close to 1 : l in all cases (Bonner et al., 1968); thus the differences observed i n protein content of these samples must be due to non-histone proteins. Positive correlations between nuclear activity and the presence of large amounts of non-histones and R K A have been observed by other workers (Dingmar? and Sporn, 1964: Marushige and Dixon, 1969; Marushige and Ozaki. 1967). Low-Salt Extraction of Chromatins: Quantitative and Qualitative Analysis. Solutions of NaC1-5 m M phosphate (pH 6.8) of ionic strength between 0.2 and 0.35 extract a substantial amount of proteins and R N A from pig cerebellum, chick embryo brain, and myeloma chromatins (Tables

~~

1684

~

~

BIOCHEMISTRY. VOL

14, ho. 8. 1975

VARIATION OF CHROMATIN CIRCULAR DICHROISM ABOVE 250 NM

+

-

--

I -

mI D

ug

I

1 1 1

u& F

0-

r

-

(A)

(C)

(E)

270

FIGURE 2: Urea-polyacrylamide gel electrophoresis of proteins from

chicken embryo brain chromatin and low salt extractable chromatin. (A) Chromatin: (E)proteins in low salt extract: [C) proteins remaining in nucleoprotein low salt. Dual gels of the same sample are loaded with different amounts of total proteins. Direction of electrophoresis is positive to negative.

290

280

300

Wovelength Inm)

FIGURE 4: Circular dichroism of chromatin extracted with Solutions of low ionic strength. (-) Chick embryo brain chromatin: (- - .-) pip cerebellum and myeloma chromatin: (. .) calf thymus chromatin and calf thymus nucleoprotein low salt: (-) pig cerebellum nucleoprotein low sale, (.) myeloma nucleoprotein low salt: (- -) chick embryo brain nucleoprotein low salt.

..

..

.~

*-

I

-

Table IV: Content of Material Removed from Chromatin by Solutions of Low Ionic Strength.

Mass Ratios (wgiwg) Sample

DNAIRNA

Pig cerebellum Chick embryo brain

1.02 * 0.12 (4)O

RoteinlRNA 47

i4

0.25 t 0.16 (9)

23

f

(5) 6 (9)

aNumber of determinations given in parentheses.

*)

+ (A)

(E)

(C)

(D)

3: SDS-polyacrylamide gel electrophoresis of proteins from chick embryo brain chromatin and low salt extracted chromatin. (A) Total histones: (B) chromatin protein; (C) proteins removed by low salt: (D) proteins remaining on nucleoprotein low salt. The boltammast band is the buffer front. The direction of electrophoresis is negative to positive. FIGURE

I1 and 111. About 50% of the R N A and 20-30% of the protein are removed from these chromatins by extraction with these solutions; however, pig cerebellum is quite variable in the amount of R N A removed (Tables 111 and IV). Calfthymus chromatin may suffer some protein loss from this treatment (Table II), but no significant loss of R N A is apparent (Table 111). The nature of the protein removed by extraction with low salt is indicated by SDS- and urea-polyacrylamide gel electrophoresis (Figures 2 and 3). Low salt extraction of chick embryo brain or pig cerebellum chromatin does not remove detectable amounts of histone proteins. Polyacrylamide gels of extracted proteins from chick embryo brain are shown as an example (Figure 2): no bands are present in the histone regions of the gels. Low salt solutions do not remove all the non-histone proteins: SDS-polyacrylamide gels of chick embryo brain samples show that the non-histone proteins removed are from the entire molecular weight range of these proteins (Figure 3). The R N A removed from chick embryo brain chromatin is found by sucrose gradient centrifugation to be smaller than 18 S. Acrylamide gels (after Loening, 1968) show the R N A to consist of discrete, low molecular weight species,

ranging in size from 4 S to 11 S. Changes in the Circular Dichroism of the Ckromatins with Low-Salt Extraction. The effects of low salt (ionic strength of 0.2-0.35) extraction on the C D of the chromatins are shown in Figure 4 and Table 1. Low salt extraction causes a considerable attenuation of the 273-nm peak in all chromatins originally demonstrating it: pig cerebellum, chick embryo brain, and myeloma (Figure I , Table I). Pig cerebellum chromatin shows the least attenuation (Table I), and perhaps this is due to the relative difficulty in removing R N A from this material (Tables 111 and IV). The C D spectra of the nucleoprotein low salt samples from these sources are now fairly similar to one another and to that of calf thymus chromatin (Figure 4, Table I). Calf thymus chromatin, on the other hand, is not significantly affected by low salt extraction. The ratios of sample ellipticities at 273 and 283 nm (Table I) verify that the shapes of the spectra for the three chromatins from pig cerebellum, chick embryo brain, and myeloma are indeed changed, whereas that from calf thymus is not. The spectra are effected by all salt concentrations between 0.2 and 0.35 M which remove non-histone and R N A . Circular Dichroism of Low Salt Extracts. With the above information on the constitution and C D of the different chromatins and nucleoprotein low salts it is possible to address the major question of this paper: what are the causes of the differences observed in the C D spectra of these chromatins? Two alternatives exist: either the material removed from chromatin by low salt possesses a sufficient C D signal to account for the observed differences; or interactions are present in some chromatins-changing the C D which are destroyed by low salt extraction. The contribution of the low salt extractable material to BIOCHEMISTRY. VOL

14. N O

8, 1975

1685

HJEL.M 1

1

'

I

I

I

I

HLANG

AKD

I

1

40t

I

301 20-

- 60

1

I

250

260

270

280

290

v

1

-201

, I

I

-25 -30-1

302

I

-

I

1

I

I '

1

I

I

,

I

4

W A V E LE N GTH

FIGURE 5 : Circular dichroism of low salt extracts from chlck embryo brain and pig cerebellum chromatins. (-) Pig cerebellum; (- - - -) chick embryo brain.

the C D above 250 nm of pig cerebellum and chick embryo brain chromatins is explored by direct observation of the low salt extracts (Figure 5). Between 255 and 275 nm the spectra of the two samples are seen to be identical with a maximum a t 265 nm of approximately 40,000 (deg cm2)/ dmol when normalized to R N A content. W e demonstrate below that this signal is sufficient to explain the observed differences among the chromatins studied. However, we need to explore some important consequences of the data given in Figure 5 before considering in detail the contribution of the low salt extractable material to the C D of chromatin. The C D spectra of the low salt extracts normalized to R N A content (Figure 5) have the remarkable characteristic that the observed ellipticity, if entirely due to R N A , is 30% greater than any reported for R N A , either isolated (Brahms and Mommaerts, 1964; Scott et al., 1968; Samejima et a]., 1968; Wolffe et al., 1968; Gratzer and Richards, 1971), or in a ribonucleoprotein particle (Sarker et al., 1967; Cox et al., 1970; Kay et al., 1970; Isenberg et al., 1971). In comparison with the C D of R N A isolated from the chick embryo brain low salt extract (about 23,000 (deg cm')/dmol R N A a t 267 nm; Figure 6) the ellipticity of the total extract is seen to be approximately 17,000 (deg cm2)/dmol of R N A greater. The relatively high salt conditions in which the C D of the low salt extract are measured d o tend to make the R N A potentially more double helical than the isolated R N A which is measured a t low ionic strength, explaining the relative blue shift seen in the samples (Gratzer and Richards, 1971), but not the extreme increase in ellipticity. Therefore, if R N A is assumed to be the only contributor to the spectra of the low salt extracts, then some heretofore unobserved conformation of R N A is involved. Alternatives exist in addition to an extremely high ellipticity of R N A in the low salt extracts. Clearly, in addition to R N A , chromophores of D N A and/or aromatic amino acids could contribute, if present. Chemical analysis of the low salt extracts of pig cerebellum and chick embryo brain chromatins (Table I V ) shows the presence of large amounts of D N A in addition to protein and R N A . The analysis indicates, also, that the pig cerebellum samples contain four times as much D N A and twice as much protein relative to the amount of R N A extracted as those of chick embryo brain chromatin. The uv spectrum of the low salt extract chick embryo brain (Figure 7) indicates the presence of rel-

1686

BIOCHE'MISTRY. VOL.

14, NO

8, 1975

atively large amounts of aromatic amino acids in the nonhistone proteins extracted. This is inferred from the maximum of the low salt extract spectrum a t 267 nm as compared to the 260 nm maximum in the uv spectrum of R N A isolated from the extract. Referring again to the C D spectra of the low salt extracts from pig cerebellum and chick embryo brain chromatins in Figure 5, another remarkable feature of the spectra is that they are identical between 255 and 275 nm. This observation in combination with the differences between the samples in protein/RNA and D N A / R N A mass ratios (Table IV) allow us to calculate the approximate contribution of D N A and protein chromophores if either do, in fact, contribute to the spectra between 255 and 275 nm. If D N A and/or protein are contributing to the spectra between 255 nm and 275 nm, then the identity of the pig and chick spectra normalized to R N A content dictates that contributions of protein and D N A within this region relative to that of R N A must be the same in both pig and chick samples. Assuming a contribution of 23,000' by the R N A to the C D of the low salt extracts, then using the DNA content of the chick samples as the lowest common denominator (Table IV), the signal of the D N A at 265 nm must be ( [ e ] R N A low salt extract - [e] isolated R N A ) C R N A / C ~ D N=A 17,000 X 4 = 68,000 (deg cm2)/dmol of DNA. This value is much higher than the highest reported values for the C D of D N A which are observed in such solvents as 80% aqueous dioxane at very low ionic strength (38,000 (deg cm2)/ dmol, Ivanov et al., 1973) and 80% ethanol in water (32,500 (deg cm')/dmol, Ivanov et al., 1973; Girod et al., 1973). Similar calculations for protein yield an ellipticity at 265 nm of +250 (deg cm2)/dmol of protein. This is a reasonable magnitude for aromatic and cystine residue-containing proteins, though a positive value for the protein side chains in this region of the spectrum is unusual (see for example, Adler et al., 1973). It should be noted that the spectra required for R N A and/or D N A to explain these results are essentially of the same position, sign, and shape as those observed for R N A (Gratzer and Richards, 1971) and D N A in the A form (Tunis-Schneider and Maestre, 1970); though, again, they differ considerably in magnitude. From these considerations there is clearly no good reason to accept any one of these possibilities over any of the other two. Each requires invoking a C D spectra that has not been observed previously. However, if large contributions from

VARIATION OF C H R O M A T I N CIRCULAR DICHROISM ABOVE 250 N M

' I '

'

'

'

'

'

'

'

'

'

I

WAVELENGTH ( n m )

FIGURE 7 : Uv spectra of low salt extract and RNA isolated from the extract of chick embryo brain chromatin. (-) Extract in 0.3 M salt; (- - - -) DEAE-isolated R N A in lo-* M NaCI. No attempt has been made to normalize the curves relative to each other.

I

,

'

250

"

I

260

"

'

270

'

"

'

280

'

J

290

300

WAVELENGTH ( n m l F I G U R E 8: Circular dichroism of nucleoprotein low salt and mathematically.constructed chromatin and native chromatin, from chick embryo brain (-): nucleoprotein low salt; low salt extract X 0.041; (- - - -) calculated chromatin CD; -) native chromatin, The error bar is plus and minus one standard deviation and applies approximately for the nucleoprotein low salt and both chromatin spectra. ( a .

- e )

(-e

D N A and proteins are to be assumed then the total contribution of D N A and/or protein to the C D between 255 and 275 nm of the low salt extracts relative to that of R N A must be the same in both the pig and chick samples. This requirement can be met if one assumes that in both the pig and chick low salt extracts the D N A and/or protein chromophores that contribute to the spectrum in this region are extracted to the same amount relative to R N A extracted. Therefore, it is attractive to speculate that the chromophores contributing to the C D of the low salt extracts are interacting in a n ordered particle. This hypothesis makes any of the above requirements plausible: constant stoichiometry and unusual CD-by way of conformations resulting from the component interactions-could both be accounted for in this manner. If the extracted material contains such a particle of R N A plus D N A and/or protein it cannot be any of the ribonucleoprotein particles which are associated with either ribosomes (removed by salineEDTA) or "informosomes" are all larger in molecular weight than the chromatin R N A s reported here (Samarina et al., 1968; Niessing and Sekeris 1970). Reconstruction of the Circular Dichroism of Total Chromatin. It remains to be demonstrated that the C D of the low salt extract observed in solution also exists to the same extent in chromatin. We have demonstrated that the C D of the low salt extracts are either due entirely to R N A or are due to R N A plus a contribution from D N A and/or protein chromophores which are extracted in a constant ratio with the R N A . In either case the C D of the low salt extract need only be multiplied by the number of decimoles of chromosomal R N A extracted/dmol of D N A in the chromatin to reconstruct the contribution of the low salt extractable material to the C D of total chromatin. In Table 111 we show that for every pg of D N A in a chick embryo brain chromatin sample 0.043 pg of R N A is removed by low salt. This is equivalent to 0.041 dmol of R N A removed/ dmol of DNA. Multiplying the C D of the chick embryo brain low salt extract by this factor, and adding the result to the spectrum of chick embryo brain low salt extracted chromatin, we obtain a spectrum that is indistinguishable from the C D of chick embryo brain chromatin (Figure 8). Thus the C D of the extractable R N A , D N A , and protein in chromatin is the same as that occurring in the low salt extract fraction. Further, this indicates that the ellipticity of the chromosomal D N A is the same before and after extraction.

Therefore, no influence on the C D of the chromosomal D N A by low salt extractable R N A and/or D N A non-histone proteins is detectable. Conclusion The evidence shows that the presence of a 273-nm peak in the C D spectra of these samples correlates best with the total amount of R N A in the sample: removal of a salt-extractable fraction containing R N A along with salt soluble non-histone proteins and D N A causes attenuation of this peak. The salt-extractable material-possibly a particleisolated from certain chromatins by solutions of low ionic strength (0.2-0.35) is present in the same form in chromatin as in the low salt extract. The low salt extract does not contain either of the ribonucleoprotein particles usually found with chromatin ribosomes, and informosomes-the size of the low salt extract associated R N A is much too small (4 S-11 S, see above). The presence of the loosely associated material in the various chromatins-as indicated by the presence of a 273-nm peak-may be closely associated with transcription activity of the chromatin. In this light it is interesting that chromatin extracted with low salt loses the capacity to synthesize R N A (Marzluff and Huang, 1974). Acknowledgments The authors thank Ms. Gloria Burgess and Mr. William Laws for their help in obtaining the experimental data. Thanks are also due to Professor Ludwig Brand for the use of the Hewlett-Packard 2100 computer, and to Professor William Harrington for the use of the Cary 60 spectropolarimeter. References Adler, A. J., Greenfield, N . J., and Fasman, G. D. (1973), Methods Enzymol. 27, 675. Bartley, J., and Chalkley, R. (1970), J . Biol. Chem. 245, 4286. Bendick, A., Russell, P. J., and Brown, G . B. (1953), J . Biol. Chem. 203. 305. Bonner, J., Dahmus, M. E., Fabrough, D., Huang, R . C. C., BIOCHEMISTRY, VOL.

14, NO. 8, 1975

1687

HJELM AND HUANG

Marushige, K., and Tuan, D. Y. H. (1968), Science 159, 47. Brahms, J., and Mommaerts, W. F. H. (1964), J . Mol. Biol. 10 73. Burton, K. (1956), Biochem. J . 62, 315. Cox, R. A., Kamgalimgan, K., and Sutherland, E. S. (1970), Biochem. J . 120, 549. DeLange, R. J., and Smith, E. L. (1971), Annu. Rev. Biochem. 40, 279. Deluca, H . A,, Rossiter, R. J., and Strickland, K. P. (1953), Biochem. J . 55, 193. Dingman, C . W., and Sporn, M. B. (1964), J. Biol. Chem. 239, 3483. Dorman, B. P., and Maestre, M . F. (1973), Proc. Natl. Acad. Sci. U.S.A. 70, 255. Elgin, S . C. R., Froehner, S., Smart, J., and Bonner, J. (1971), Adv. Cell Mol. Biol. 1 , 1. Fairbanks, G., Theodore, L., and Wallach, D. F. H . (1971), Biochemistry 10, 2606. Georgiev, G . P. (1969), Annu. Rev. Genet. 3, 155. Girod, J. C., Johnson, W . C., Huntington, S. K., and Maestre, M. F. (1973), Biochemistry 12, 5092. Gordon, D. J. (1972), Biochemistry 1 1 , 413. Gratzer, W. B., and Richards, E. G. (1971), Biopolymers 10, 2607. Hearst, J . E., and Botchan, M. (1970), Annu. Rev. Biochem. 39, 151. Hjelm, R. P., and Huang, R. C. (1972), Biophys. J. 12, 246a. Hjelm, R. P., and Huang, R . C . (1974a), Biochemistry 13, 5275. Hjelm, R. P., and Huang, R. C. (1974b), manuscript in preparation. Holzwarth, G., Gordon, D. G., McGinniss, J. E., Dorman, B. P., and Maestre, M. F. (1974), Biochemistry 13, 126. Huang, R. C., and Hjelm, R . P. (1974), in Handbook of Genetics, Vol. 5, King, R., Ed., New York, N.Y., Plenum Publishing Co., (in press). Huang, R . C., and Huang, P. C. (1969), J . Mol. Biol. 39, 365. Isenberg, H., Cotter, R. I., and Gratzer, W . B. (1971), Biochim. Biophys. Acta 232, 184. Ivanov, V. I., Minchenkova, L. E., Schyolkina, A . K., and Poletayev, A. I. (1973), Biopolymers 12, 89. Kay, C. M., Colter, J. S., and Oikawa, K. (1970), Can. J .

1688

BIOCHEMISTRY, VOL.

14, NO. 8 , 1 9 7 5

Biochem. 48, 940. Koenig, H. (1958), Proc. SOC.Exp. Biol. Med. 97, 255. Laemmli, U. K. (1970), Nature (London) 227, 650. Leblond, C. P., and Walker, B. E. (1956), Physiol. Rev. 36, 255. Loening, U. E. (1 968), J. Mol. Biol. 38, 355. Lin, J. C., Nicoloni, C. and Baserga, R. (1 974), Biochemistry 13, 4127. Lowry, 0. H., Rosebrough, N. J., Farr, A . L., and Randall, R. J. (1951), J.Biol. Chem. 193, 265. Marushige, K., and Dixon, G. (1969), Dev. Biol. 19, 397. Marushige, K., and Ozaki, H . (1967), Dev. Biol. 16, 474. Marzluff, W . F., Jr., and Huang, R. C. (1974), Proc. Natl. Acad. Sci. U.S.A. (in press). Murray, K., Vidali, G., and Neelin, J . M . (1968), Biochem. J . 107, 207. Neelin, J. M., Callahan, P. X., Lamb, D. C., and Murray, K. (1 964), Can. J. Biochem. 42, 1743. Niessing, J., and Sekeris, C . E. (1970), Biochim, Biophys. Acta 209, 484. Panyim, S., and Chalkley, R . (1969), Biochemistry 8, 3972. Samarina, 0. P., Lukanidin, E. M., Molnar, J., and Georgiev, G. P. ( 1968), J . Mol. Biol. 33, 25 1. Samejima, T., Hashizuma, H., Iamahori, K., Fujjii, I., and Miura, K. (1 968), J. Mol. Biol. 34, 39. Sarker, P. K., Yang, J . T., and Doty, P. (1967), Biopolymers, 5, 1. Schneider, W. C. (1957), Methods Enzymol. 3, 680. Scott, J. F., Monier, R., Aubert, M., and Reynier, M. (1968), Biochem. Biophys. Res. Commun. 33, 794. Shaw, L. M .J., and Huang, R. C. (1970), Biochemistry 9, 4530. Shih, T. Y . , and Fasman, G. D. (1970), J . Mol. Biol. 52, 125. Simpson, R . T., and Sober, H . A. (1970), Biochemistry 9, 3103. Stellwagen, R . H., and Cole, R. D. (1969), Annu. Rev. Biochem. 38, 95 1. Tunis-Schneider, M . J., and Maestre, M . F. (1970), J . Mol. Biol. 52, 52 1 . Vidali, G., Boffa, L. C., Littau, V. C . , Allfrey, K. M., and Allfrey, V. G . (1973), J . Biol. Chem. 248, 4065. Williams, A. F. (1971), J . CellSci. 10, 27. Wolffe, F. H., Oikawa, K., and Kay, C. M . (1968), Biochemistry 7 , 3361.